Five known melanosome-specific proteins currently serve as immune targets for melanoma, i.e. TYR, TRP1/gp75, TRP2/Dct, Pmel17/gp100 and MART1. Host cellular and humoral immune responses play important roles in autoimmune pigmentary diseases (e.g. vitiligo) but are usually unable to control the growth of primary melanomas or their metastases. We are attempting to identify novel melanocyte-specific proteins expressed by less differentiated melanocytic cells that should provide more immune targets for diagnosis and/or therapy of melanoma. We are also looking at approaches to regulating the expression of genes in melanocytic cells which modulate their growth, and are mapping loci that are closely linked to the growth and metastases of melanomas in various locations on the body. Molecular approaches to clone pigment-related genes have been highly successful, but despite that only 50% of the greater than 200 known pigment-related genes have been cloned leaving many potential melanocyte-specific targets to be identified. Tissue microarray and proteomics approaches to define markers of melanoma progression have also identified useful tumor markers. We have used a more direct approach, i.e. to purify melanosomes and use mass spec microsequencing to identify their protein constituents. We identified 1,500 proteins in melanosomes of all stages of maturation, with 600 in any given stage;they include 16 homologous to mouse coat color genes and many associated with human pigmentary diseases. As a result, we successfully published online the first global proteome database, an early gold standard for future proteomics studies on melanosomes. This project is now focused on identifying novel and specific melanosomal proteins that may prove to be useful targets for the immunodiagnosis and/or immunotherapy of melanoma. We are examining differences in proteins expressed in various stages of melanosomes produced by pigmented and/or by amelanotic melanoma cells. Special emphasis will be made to identify new markers potentially suitable to identify amelanotic melanomas, which are difficult to detect and are highly aggressive. We have also identified the processing and trafficking of GPNMB as another potential marker for melanoma cells. Our research has focused on one of the most frequent causes for failure to detect highly aggressive melanomas, which is the lost expression of melanoma antigens. We have found that melanoma cells lose expression of critical melanoma antigens such as Pmel17/gp100. Further, our work has revealed that Pmel17 expression and processing is tightly linked to the growth of melanoma cells via other critical intracellular signaling intermediates. We have identified SOX9 as a critical transcription factor involved in regulating melanocyte function and proliferation, which is down-regulated in a large number of melanoma cells. Up-regulation of SOX9 expression in melanoma cells is an effective way to inhibit their growth and to decrease their metastatic potential. We have identified approaches to up-regulate SOX9 expression using existing pharmaceutical agents that not only increases SOX9 expression, but increases the sensitivity of melanoma cells to retinoic acid, which might prove to be an effective way to target tumors in melanoma patients. We are collaborating with Dr. Gottesman's MDR group to define yet another unexpected function of melanosomes, which involves the specific expression of an MDR protein (ABCB5) in melanocytes and its localization in melanosomes. MDR mechanisms underlying the intractability of melanomas to chemotherapy remain largely unknown but are a critical therapeutic challenge. We found that MDR (at least in some melanomas) involves the sequestration of cytotoxic drugs within subcellular organelles (including melanosomes), which significantly reduces their nuclear localization and efficacy. Our study provides evidence that melanosomes contribute to the refractory nature of melanoma cells by sequestering cytotoxic drugs and increasing melanosome-mediated export of drugs. Preventing sequestration of cytotoxic drugs by inhibiting the functions of melanosomes may have great potential as an approach to improve the chemosensitivity of melanoma tumors. We hypothesize that ABCB5 plays an important role in normal melanosomes to ensure that cytotoxic intermediates of melanin synthesis remain within that organelle, which are eventually lost by desquamation of the skin or growth of hair, and we are testing that hypothesis. Regarding our studies of pigmentary disorders, mutations in the TYR gene result in OCA1, the most dramatic and severe form of that disease, where little or no melanin is produced in the hair, skin or eyes. Mutations in the gene encoding TYRP1 affect its interactions with TYR, and result in the hypopigmented phenotype of OCA3. Mutations in Tyr or Tyrp1 elicit the retention of TYR in the ER and its degradation by proteasomes, revealing that OCA1 and OCA3 are in fact ER-processing diseases. OCA2 and OCA4 are relatively severe forms of OCA which arise from mutations in the P and MATP loci, respectively. Both encode highly similar proteins with 12 transmembrane motifs. We found that both OCA2 and OCA4 result from altered trafficking of TYR after processing in the ER and Golgi en route to melanosomes, which reduces pigmentation despite the presence of functionally active TYR. We are testing the hypothesis that P and MATP affect the pH of sorting vesicles leading to disrupted trafficking of TYR to melanosomes. Our studies on the disrupted processing and sorting of TYR in OCA2 and in OCA4 melanocytes explains their hypopigmented phenotypes will provide new insights into the involvement of transporters in the normal physiology of melanocytes. Mutations in sorting determinants also result in a hypopigmentary disease known as HPS, which is due to the misrouting of many cellular components (including melanocyte-specific proteins) and thus has effects on other types of tissues as well as pigmented ones. We are characterizing novel forms of HPS mutations and the mechanisms of disrupted trafficking in mouse melanocyte models and are in the process of characterizing the functions of P and MATP.